Please use this identifier to cite or link to this item:
Title: Calibration of stochastic computer models
Authors: YUAN JUN
Keywords: Stochastic computer model,computer model calibration,stochastic approximation,parameter uncertainty,Gaussian process,sequential experimental design
Issue Date: 5-Aug-2013
Citation: YUAN JUN (2013-08-05). Calibration of stochastic computer models. ScholarBank@NUS Repository.
Abstract: This thesis studies the calibration of stochastic computer models. When a computer model is used to predict the behavior of a real system for decision making, it is important to calibrate the computer model so as to improve the model?s predictive accuracy. This thesis first proposes an automated calibration approach for stochastic computer models based on the stochastic approximation method that can search for the optimum calibration parameter values accurately and efficiently. In order to use the limited data resources more efficiently when computer models are extremely time consuming, and better quantify the various uncertainties, this thesis next proposes a surrogate based Bayesian approach for stochastic computer model calibration and prediction. Numerical results show the accuracy and efficiency of the proposed Bayesian calibration approach. Furthermore, in order to effectively allocate the limited data resources, a general two-stage sequential approach is proposed for stochastic computer model calibration and prediction. Several different design criteria are proposed for the selection of points in the sequential approach, and several examples are used to compare their calibration and prediction performances. Finally, an overall general framework is provided to combine the calibration, validation and prediction processes for stochastic simulation. Based on this framework, an integrated approach is proposed for stochastic computer model calibration, validation and prediction to facilitate the development and use of stochastic computer models.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
YuanJ.pdf2.44 MBAdobe PDF



Page view(s)

checked on Oct 25, 2020


checked on Oct 25, 2020

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.